Doing Good Business by Hiring Directors with Foreign Experience

On September 11, 2008, the Gansu Health Bureau released an announcement that Sanlu Corporation, one of the largest dairy producers in China, had placed products contaminated by melamine on the market. Three critical consequences resulted from this contamination. First, more than 290,000 people (most of them children) were poisoned, leading to the deaths of six infants (Kong 2012). Second, twenty-two dairy producers in China, with total market shares of over 90% in liquid milk and over 50% in powdered milk, failed quality examinations due to melamine contamination (Dai et al. 2013). Third, the New Zealand dairy cooperative Fonterra, which comprised business partners and shareholders of Sanlu Group, wrote down the carrying value of its investments by NZ$ 139 million. In fact, the top managers of these dairy producers knew that their products contained melamine harmful to humans, but they were concerned about the loss of short-term profits and the damage to firm reputation if they recalled the products from the market. Currently, most Chinese families prefer milk products imported from New Zealand, Holland, and Japan. Therefore, Chinese milk companies lost their market shares and experienced financial distress.

One anonymous referee raises the concern that director turnover may challenge our finding because turnover effectively changes the returnee-director ratio. Due to the limited sample observations, we did not filter the firms with director turnover. To alleviate the concern raised by the referee, we conduct a robustness check in an unreported regression. In the robustness check, we define a new dummy variable to indicate whether a firm experiences director turnover during the sample period. Then, we add the interaction term between overseas director ratio and the dummy variable of director turnover to the regression. The result shows that the coefficient of the interaction term is insignificant and that the coefficient of overseas director ratio remains positive and highly significant. This result suggests that our finding is not affected by director turnover. These results are available upon request.

We remove the control variable for government ownership from the subsample analysis when we split the sample firms into with- and without-government-ownership subsamples. We also remove the control variable for leverage when we split the sample firms into high- and low-leverage subsamples. We thank the anonymous referee for this comment.

In Table 8, we measure whether a firm has been included in the Running’s Report. Thus, the sample firms include all public firms listed on the Chinese stock market. This differs from previous test samples, which comprise only firms included in the Running’s report.